This is a renewal of a project to develop (and share with other investigators) a highly informative resource of bipolar, type I (BPI) families to test for linkage to marker loci spanning the human genome and to candidate genes. In order to achieve our purposes we continue to 1) examine families, ascertained through a proband with BPI with two or more affected sibs (or with one affected sib and one affected parent); 2) establish lymphoblastoid lines on all participating family members; 3) analyze the variations in clinical features between and within families; and 4) employ informative genetic markers spanning the human genome in order to conduct linkage analyses. The clinical investigators at Johns Hopkins, assisted by consultants at the University of Iowa, are conducting an intensive screening program to ascertain only highly suitable, unilineal bipolar families. After two years of study, over 1600 BPI families have been screened and rejected. Of the remaining 96 families, 17 unilineal BPI families are suitable and are completely clinically assessed. We anticipate completing 25 such families by the end of Year 3 of the original grant. We are asking for funds in this renewal proposal to ascertain and study sufficient families to yield another 25 unilineal BPI families, and to complete DNA studies, linkage analyses, and clinical analyses on a final set of 50 unilineal BPI families. We will use a systematic RFLP linkage map of the human genome, probes for candidate loci, newer PCR-derived probes, and karyotypes to search for relevant loci. Our consultant, Dr. Eric Lander, has developed newer methods for linkage analysis for complex disorders, which we will employ in addition to more standard linkage analyses.